Finnish Hospital Infection Program (SIRO), National Institute for Health and Welfare (THL), Department of Infectious Disease Surveillance and Control, Helsinki, Finland.
J Antimicrob Chemother. 2011 Nov;66(11):2651-4. doi: 10.1093/jac/dkr333. Epub 2011 Aug 16.
It is difficult to draw conclusions about the prudence of antibiotic use in different hospitals by directly comparing usage figures. We present a patient case-mix adjustment model of antibiotic use to rank hospitals while taking patient characteristics into account.
Data on antibiotic use were collected during the national healthcare-associated infection (HAI) prevalence survey in 2005 in Finland in all 5 tertiary care, all 15 secondary care and 10 (25% of 40) other acute care hospitals. The use of antibiotics was measured using use-days/100 patient-days during a 7day period and the prevalence of patients receiving at least two antimicrobials during the study day. Case-mix-adjusted antibiotic use was calculated by using multivariate models and an indirect standardization method. Parameters in the model included age, sex, severity of underlying diseases, intensive care, haematology, preceding surgery, respirator, central venous and urinary catheters, community-associated infection, HAI and contact isolation due to methicillin-resistant Staphylococcus aureus.
The ranking order changed one position in 12 (40%) hospitals and more than two positions in 13 (43%) hospitals when the case-mix-adjusted figures were compared with those observed. In 24 hospitals (80%), the antibiotic use density observed was lower than expected by the case-mix-adjusted use density.
The patient case-mix adjustment of antibiotic use ranked the hospitals differently from the ranking according to observed use, and may be a useful tool for benchmarking hospital antibiotic use. However, the best set of easily and widely available parameters that would describe both patient material and hospital activities remains to be determined.
通过直接比较使用数据，很难对不同医院使用抗生素的谨慎程度得出结论。我们提出了一种患者病例组合调整的抗生素使用模型，以便在考虑患者特征的情况下对医院进行排名。
2005 年在芬兰进行的全国医疗相关感染（HAI）患病率调查期间，在所有 5 家三级保健医院、15 家二级保健医院和 10 家（40 家的 25%）其他急性保健医院中收集了抗生素使用数据。使用 7 天期间的使用天数/100 患者天数来衡量抗生素的使用情况，并测量研究日至少接受两种抗菌药物的患者比例。通过使用多变量模型和间接标准化方法计算病例组合调整后的抗生素使用情况。模型中的参数包括年龄、性别、基础疾病严重程度、重症监护、血液科、手术前、呼吸机、中心静脉和导尿管、社区相关感染、HAI 和耐甲氧西林金黄色葡萄球菌的接触隔离。
与观察到的情况相比，当比较病例组合调整后的数字时，12 家（40%）医院的排名发生了一个位置的变化，13 家（43%）医院的排名发生了两个以上的变化。在 24 家（80%）医院中，观察到的抗生素使用密度低于病例组合调整后的使用密度。
对抗生素使用的患者病例组合调整改变了医院的排名，与观察到的使用情况不同，这可能是基准医院抗生素使用的有用工具。然而，描述患者资料和医院活动的最佳、容易和广泛获取的参数集仍有待确定。
